Influenza is a leading cause of death and morbidity in the US, resulting in up to 40,000 deaths, ~200,000 hospitalizations and > $10.4 billion in expenses per year. Current antivirals incur high resistance rates and have low efficacy so there is a crucial need to develop better approaches to protect from seasonal influenza and future pandemics. We aim to develop computationally designed peptides as a pre-exposure prophylactic and post-exposure therapeutic against influenza. We have shown that a small computationally designed protein (HB36.6), interferes with influenza infection by binding the highly conserved hemagglutinin stem region and neutralizes a wide range of influenza variants including pandemic strains and strains resistant to current antivirals. When delivered intranasally, this influenza binder confers complete prophylactic and therapeutic protection against different influenza strains in mice, a result that demonstrates for the first time, transformation of a theoretical target designed by this computational method into a new antiviral with strong biopotency in vivo. In this application, we propose to build on our promising findings with HB36.6 and broaden efficacy by computationally designing de novo hyperstable mini-protein binders (30-40 amino acids) that will be effective against influenza providing pan-specific protection against a broader range of influenza strains. These new hyperstable mini-binders, will be smaller, more stable, have increased bioavailability, and will have a lower cost of manufacturing than our current lead Group 1 binder HB36.6. We hypothesize that the optimized computationally designed mini-protein binders will afford superior broad, potent prophylactic and therapeutic protection in mice against Group 1 viruses. We will determine expression and in vitro neutralization of computationally designed mini-proteins and identify a single mini-protein binder that affords the best prophylactic and therapeutic protection in mice challenged with a lethal dose of Group 1 influenza virus. If successful, these experiments will establish a new class of flu antivirals that could overcome the limitations of currently marketed flu antivirals an build a strong preclinical data package for future development.